sign in sign up
<<
Home , Cima volcanic field, Cinder cone, Southern Cone, Volcanoes of Kamchatka

Journal of Volcanology and Geothermal Research 201 (2011) 301–311

Contents lists available at ScienceDirect

Journal of Volcanology and Geothermal Research

journal homepage: www.elsevier.com/locate/jvolgeores

Morphometric and morphological development of cinder cones: A field and remote sensing study in the volcanic field, Kamchatka

Moshe Inbar a,⁎, Michael Gilichinsky b, Ivan Melekestsev c, Dmitry Melnikov c, Natasha Zaretskaya d a Department of Geography and Environmental Studies, University of Haifa, Israel b Department of Forest Resource Management, SLU, Umea, Sweden c Institute of Volcanology and Seismology, RAS, Petropavlovsk-Kamchatsky, d Geological Institute, RAS, Moscow, Russia article info abstract

Article : The evolution of landscape over time is a central aspect of geological, paleogeographical and Received 14 January 2010 geomorphological studies. Volcanic features like cinder cones offer the opportunity to monitor the processes Accepted 18 July 2010 and development of the landscape. Cinder cones are perhaps the simplest and most common volcanic Available online 24 July 2010 landforms in the world. Morphological and morphometric study of cinder cones has proven an efficient tool for determining their relative dates, and the erosional processes affecting them. The extensive Kamchatka Keywords: volcanic province (Russian ), with its large Tolbachik field, is an excellent case study for Kamchatka volcanic province fi Tolbachik eruption spatial and temporal classi cation and calibration of changes in morphometric values with time. cinder cones We show how the morphological and morphometric values of the monogenetic cinder cones, measured in morphometry the field and by digital elevation models, can be used to validate their age and erosional processes. volcanic geomorphology Field data were GPS measurements of cinder cones formed at the Tolbachik 1975–1976 eruption and of landscape evolution Holocene cinder cones; processes on the cinder cones and flows were identified and evaluated. For every studied cinder cone morphometric parameters were assessed on the basis of remotely sensed data and digital elevation model. Morphometric measurements were taken of cone height and slope and average axis diameter and the height–width ratio was obtained. The comparison of morphometric parameters calculated from ASTER DEM and topographic map clearly supports the concept of relative morphometric dating as the most recent cinder cones are always associated with the highest slopes and h/W ratio. The measured morphometric values of the recent Tolbachik cinder cones are valuable benchmark data for determining erosion rates, such as the measured values for the Paricutin cone in after the 1943 eruption. The variability of the morphometric values of the recent cinder cones is due to their lithological coarse composition. A comparison with the older cinder cones in the area shows that the climatic conditions of the Kamchatka and the slow development of vegetation cover determine a high rate of erosion and rapid change in the morphometric values, as compared to published values for other volcanic fields. © 2010 Elsevier B.V. All rights reserved.

1. Introduction and morphometric study of cinder cones has proven an efficient tool for determining their relative dates of cinder cones and the erosional Morphological studies of volcanic landforms have multiplied in processes affecting them (Wood, 1980b; Hasenaka and Carmichael, recent decades, primarily after the eruption of St Helens in 1985; Hooper, 1995; Inbar and Risso, 2001; Parrot, 2007). Such 1980 and in the light of the need to interpret extraterrestrial analyses generally attempt to measure the morphometric values of morphologies (Wood, 1980a; Thouret, 1999). cinder cones, e.g. height, diameter, and slope, and the relations among Cinder cones are “perhaps the simplest and most common volcanic them. The evolution of cinder cones erosion might be associated to the landforms in existence” (Wood, 1980b). It is probably the only one on period of time of exposure to degradation processes. The progressive the globe with a distinct and defined initial date of formation, and change of morphometric parameters with increasing of age is the lasting no more than a few million years, before erosional processes basis for relative dating of cones by comparative measurements flatten it, a short time in the 's geological history. Morphological (Wood, 1980a; Hooper and Sheridan, 1998). The evolution of landscape over time is a central aspect of geological, paleogeographical and geomorphological studies. Recent ⁎ Corresponding author. volcanic features like cinder cones offer the opportunity to monitor E-mail address: [email protected] (M. Inbar). the processes and development of the landscape. Morphometric and

0377-0273/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.jvolgeores.2010.07.013 302 M. Inbar et al. / Journal of Volcanology and Geothermal Research 201 (2011) 301–311 morphological studies, together with remote sensing, tephrochronol- 2. Physical background ogy and methods of absolute dating, are efficient tools for determining ages of cinder cones, their morphological evolution and spatial–time Kamchatka forms the northern segment of the Kurile-Kamchatka development of aerial volcanic fields all over the world and in island arch which is a part of Pacific “fire ring.” This peninsula extends extraterrestrial conditions. for ca 1300 km, SW to NE between 51° and 62° N. Kamchatka and the The large Kamchatka volcanic field (), with adjacent Kurile Islands are the above-water part of a 2000-km probably the largest number of monogenetic cinder cones in the asymmetrical range, separating the of Okhotsk from the Pacific world, is an excellent case study for spatial and temporal classifica- (Ponomareva et al., 2001), underwater part is 300–700 km tion. Many zones of monogenetic volcanism exist, with tens of cinder wide; the maximum width of the landmass is 420 km. Main property cones (such as the Tolbachinsky, Tolmachev and Sedanka fields), and of the last 100 million years of Kamchatka geological history is the fields of parasitic cones on the slopes of large volcanoes (such as continuous magmatic and volcanic activity proceeding at varying Kliuchevskoy plateau). Cones are located all along the ~1000 km of intensity over time. For most of the Cenozoic period an orogenic the Kamchatka , at various elevations above sea level; foldbelt was developed, and in the Pliocene– a volcanic they are of different types of pyroclastic components. In particular, arch developed with adequate volcanic processes and landforms. they are located at various distances from large volcanic centers, that The modern climate is very humid in most of the peninsula, with is, they are covered by soil-pyroclastic successions of different precipitation rate ranging from 2500 mm/year on the east coast to thickness, and this influences their morphometric features. 500–800 mm/year in the central part of Kamchatka. Winters are cool , like many other volcanic areas of the world, and snowy and summers are wet and cloudy (Solomina et al., 2007). is blanketed by a soil-pyroclastic cover, representing the alteration of The Tolbachik cinder cone field is one of the youngest volcanic ashes of different volcanoes and buried soils (peat) or sandy loams. landforms in Kamchatka, located in the SW part of the Kliuchevskaya This cover is a few tens of centimeters thick in areas far from the active volcanic group, which is the largest in Kamchatka (Fig. 1). The volcanoes and increases up to several meters at their source. The Tolbachik field is a huge, inclined lava plane, formed as a result of continuously accumulating soil/peat-pyroclastic successions in the abundant eruptions of many cinder and lava cones and some fissure areas under the Late-Pleistocene glaciation are usually of Holocene eruptions, with a total surface area of 875 km2 and these volcanic ages; older covers have been eroded by series of glaciations. centers are related to the 70-km long Tolbachik regional cinder cone The Late Pleistocene and Holocene cinder cones are well preserved zone. This zone has a NNE bearing (azimuth 20°–25°) in the in Kamchatka while older forms (e.g. of Early–Middle Pleistocene age) crossing the Plosky Tolbachik volcano, and changing direction to NE are markedly eroded due to geomorphic processes, particularly (azimuth 40°–45°) (Fig. 1). during the last stage of the Pleistocene glaciation. Unlike other large volcanic fields with hundreds of monogenetic 3. Tolbachik — chronology of eruptions cinder cones, such as the Michoacan-Guanajuato (Mexico), San Francisco (, USA), Andino-Cuyano (Mendoza, ), The Tolbachik cinder cones erupted during the last ten kiloyears Tenerife (Canary Islands, Spain) and with a long list of (ky) at 200–2500 m a.s.l. The lower time boundary estimation was published works (Scott and Trask, 1971; Porter, 1972; Bloomfield, based on the evidence that the moraine of the last Upper Pleistocene 1975; Settle, 1979; Martin del Pozzo, 1982; Dohrenwend et al., 1986; glaciation was covered by lava and pyroclastic deposits of the oldest Wolfe et al., 1987; Hooper, 1995; Inbar and Risso, 2001; Dóniz et al., cinder cones. The age-related cinder cone groups were subdivided 2008, among others), the Kamchatka cinder cone field enjoys few based on tephrochronological studies together with extensive morphological studies (Syrin, 1968; Dirksen and Melekestsev, 1999); . The following cinder deposits of catastrophic no detailed morphometric work at all has been published on this volcanic eruptions were used as marker ash layers (Braitseva et al., volcanic province. 1993): Young Shiveluch, Ksudach, Khangar and Kizimen. As a result, The introduction of digital elevation models (DEM) as digital six age-graded groups were defined: representation of surface topography has made precise quantitative estimation of cinder cone characteristics, possible. Parrot (2007) (1) 10–7.5 ky presented a method for accurate calculation of various geomorphic (2) 7.5–2ky parameters from high-resolution DEMs (Chichinautzin Range Volca- (3) 2–1.5 ky nic Field, Mexico) to simulate the evolution of a given shape and then (4) 1.5–1ky to estimate the volume of material removed during erosion. Kervyn et (5) 1–0.3 ky al. (2008) performed a comparative analysis of the accuracy of DEMs (6) Recent eruptions (1740, 1941 and 1975–1976) provided by ASTER (Advanced Space-borne Thermal Emission and Reflection Radiometer) and SRTM (Shuttle Radar Topographic The last, huge eruption of 1975–1976 was the youngest volcanic Mission), focusing on retrieval of quantitative morphometric data episode of the Tolbachik field activity during the Holocene; and there for moderate and small-sized volcanic features. were ca 60 volcanic episodes during the last 10–11 ky. During the last The aims of this study were: a. to examine the morphology and 2 ky large cinder cones were formed, composed of magnesia , morphometric relationship among the monogenetic cones in the like the northern cones of the GFTE. Cones formed earlier are far Kamchatka cinder cone field, with measuring to be done mostly in the smaller in size and volume. Despite the short-term span of cinder cone field; and b. to analyze their relative ages as compared with absolute existence (only 10–11 ky) their surface morphology (especially of the dates. This would allow a new interpretation of the sequence oldest cones) notably changed (Fig. 2). development of the volcanic field, as well as the rate of erosional processes in the climate of the Kamchatka Peninsula compared with 4. The 1975–1976 eruption volcanic fields in different climates in the world to examine the consistency of morphometric characteristics estimated from digital The Great Tolbachik Fissure Eruption (1975–1976) is the youngest elevation models (DEM) of nine monogenetic cones within the eruption of the Tolbachik cinder cone field, when cinder cones of Tolbachik volcanic field. Four “new” cinder cones were recently various in dimension, morphology and chemical composition formed. formed during the Great Fissure Tolbachik Eruption (GFTE) of The eruption started on June 28, 1975 from weak explosions from the 1975–1976 and five “old” cones were formed during the last Plosky Tolbachik summit crater. On July 6, the Northern cones started 1700 years. to form. M. Inbar et al. / Journal of Volcanology and Geothermal Research 201 (2011) 301–311 303

Fig. 1. Location map of the study area and overview of Kamchatka peninsula. Northern and Southern fissures of Great Fissure Tolbachik Eruption (GFTE) are depicted in dark grey on the grey background of Holocene lava flows. Adapted from Fedotov and Masurenkov, 1991.

During the 72 days of Northern Fissure activity, three cinder cones 1975 the parameters of the cinder cones were measured from aerial formed, together with a lava field (S=8.86 km2) with a lava cover photo data: I — height 299 m and volume 0.133 km3;II— 278 m 80 m thick (northern dark grey area in Fig. 1). At the end of August height and volume 0.099 km3; and III — 108 m height and volume

Fig. 2. View of Kamenistaya old cone (left) and cone Yuzhniy (right) erupted in 1976. Slope values are 23° and 31° respectively. 304 M. Inbar et al. / Journal of Volcanology and Geothermal Research 201 (2011) 301–311

Fig. 3. Perspective view of the study area presented by ASTER DEM.

0.022 km3 (Dvigalo et al., 1980). Their total volume was estimated at pyroclastic rocks (0.921 km3 and 1.117×109 tons) formed the cinder 0.254 km3 and the weight of composing rocks at 0.38×109 tons cones. The respective mean rates of pyroclastic outflow were 178 m3/s (ρ=1.5 g/cm3). The primary slope angle of the cones was 30°–35°. and 214 tons/s. The total volume of was 0.223 km3 and the Only 28% of total volume and 34% of total weight of erupted weight was 0.49×109 tons (ρ=2.2 g/cm3). The lavas and pyroclastic rocks of the Northern Fissure had similar chemical composition and are mostly magnesia medium alkaline

basalts (49.54–49.75% of SiO2, 9.84–10.21% MgO, 2.38–2.44% Na2O, 0.97–1.02% K2O) according to 35 samples analyzed (Braitseva et al, 1997). Remarkably, the previous eruption which took place in 1941 had similar eruption dynamic, parameters (i.e. rate of pyroclasts and lava volume) and chemical composition to those of the Northern Fissure eruption. We can reasonably assume that the above peculiarities of magnesia cinder cone formation and the dynamic parameters of their eruption had been characteristic for the older cinder cones which started forming at the Tolbachik field ca 2000 BP (Braitseva et al., 1997). The Southern cone appeared 10 km SW of Cone I of the Northern Fissure at 380 m a.s.l. and erupted ca 450 days (6 times more than the Northern Fissure). Its activity was quieter version of Strombolian-type eruption. A single relatively small cone formed: its maximum height was 165 m, rock weight — 0.024×109 tons (ρ=1.5 g/cm3), the volume — 0.016 km3 (together with cone fragments dragged down by lava flows). The relative height of the cone (after the eruption ceased) was 85–120 m (its base was buried under lava flows 80 m thick). The main result of the Southern cone eruption was the formation of a huge lava cover (S=35.87 km2, V=0.968 km3 and weight=2.14×109 tons). The moderate and weak explosive activity of the Southern cone caused most of pyroclastics (53%) to accumulate on the cone itself: 47% was blown away by the wind. The lavas and pyroclastics of the South cone consist of subalkalic, aluminous basalts (50.67–50.85%

SiO2, 4.58–4.98% MgO, 3.52–3.62% Na2O, 1.85–2.11% K2O, 133 samples analyzed) (Flerov et al., 1984).

5. Methodology

5.1. Field measurements

In our fieldwork in August 2007 on the Tolbachik cinder cone field (GTFE area) we studied the morphometry of the nine cinder cones located along the main sub–meridian fault (Fig. 3). The cones lie at various altitudes (350–1500 m a.s.l.), and the distance between the Fig. 4. Field measured elevation profiles of recent (a) and old studied cinder cones (b). Note that height is given as absolute elevation above sea level ground (photo August outermost Yuzhniy cone on the south and the Krasny cone on the 2007). north is 20 km (Fig. 3). M. Inbar et al. / Journal of Volcanology and Geothermal Research 201 (2011) 301–311 305

Fig. 5. Trees burnt during the 1975 eruption. Grass and moss vegetation developed on ground (photo August 2007).

The elevation measurements were obtained as WGS84 ellipsoid (h/W), where h is the relative height and W=(dmj +dmi)/2. We took heights by a Fujitsu–Siemens Pocket LOOX N560 (Sirf III GPS chipset). field GPS measurements (plane coordinates and elevation) of 40 This handheld GPS device provides a horizontal accuracy of about 3 m locations to provide geodetic reference for DEM processing and also to and its error in elevation can be as high as 7–9 m. In total we collected validate the calculated DEMs at additional 45 locations (Fig. 8). GPS data at 85 locations within the study area including 73 measurements of randomly oriented elevation profiles of the nine 5.2.1. Topographic maps cinder cones. We thus obtained a series of elevation profiles based on Cone height (h), cone width (W), h/W ratio and slope angle 8–10 observation points each profile (Fig. 4). calculated from topographic maps are the main morphometric The ground control points (GCP) were corrected to geoid heights parameters used for determining the rate of cone degradation as a (EGM96) and used in DEM generation, employing the DEM Extraction function of its exposure to erosion (Wood, 1980a, b). The standard Module (ENVI™ 4.6). The GPS measurements were also used to definition of these parameters is based on measurements derived evaluate the accuracy of the DEMs generated from ASTER remotely from topographic maps. In Hooper and Sheridan (1998) cone width is sensed data. calculated as the average of the maximum and minimum basal The recent cones are surrounded by a covered plain where diameters for each cone and cone height is defined as the difference vegetation cover is about 10% and most of it are mosses. No vascular between average basal elevation and maximum crater rim or summit plants were found in the new cones. Trees in the affected area by the elevation. The methodology adopted to measure the morphometric eruption were completely burned (Fig. 5) and in the cones area no parameters over topographic maps is similar to that introduced by burnt tree trunks remained. Most of the new cinder cones surround- Wood (1980a) and summarized by Hooper (1995). Cone basal ing area is covered by and lava, on the base of the cones diameter was measured from topographic maps and calculated as megablocks were found about one meter diameter (Figs. 6 and 7). the mean of the maximum and minimum basal diameters of each Depth of craters in the recent cones is about 50 m (Fig. 6) and they cone. Cone height was measured as the difference between the are unbreached in all the cones, except Cone N2. In all cones slopes are highest and the average basal elevation. steep and built of about 50% of lapilli and scoria material of the surface The mapping of erupted material from four cinder cones was slope. They are not dissected except for trails of big volcanic bombs. In carried out already during GFTE whereas the entire Tolbachik N2 “bocca lava” was found close to the crater that breached the field was topographically surveyed at a scale of 1:25,000 in cone flank. The old cones are eroded by breaching with broad tails in 1976–1980. In this study we used topographic maps of the Tolbachik their base, as evidence of mass wasting processes. volcano field on a scale of 1:25,000 were used for calculation of morphometric parameters and for providing comparable basis with DEM-based morphometric measurements (Fig. 9a and b). 5.2. Morphometric measurements

Morphometric measurements from the DEM derived from ASTER 5.2.2. ASTER DEM optical imagery with 30 m of spatial resolution were taken, and these The present research uses remotely sensed imagery of ASTER were compared with measurements done on 1:25,000 topographic (Advance Space-borne Thermal Emission and Reflection Radiometer). maps. For every studied cinder cone we calculated the main This is cloudless (March 2004) Level 1A ASTER image products morphometric parameters from these two elevation data sources. (Reconstructed Unprocessed Instrument Data V003) and orthorecti- They include cone height (h), steepest cone slope, major-axis fied Level 3 images. The visible near infra-red sensor of Level 1A diameter (dmj), short axis diameter (dmi), and height–width ratio ASTER products (with 15 m of spatial resolution) includes nadir- 306 M. Inbar et al. / Journal of Volcanology and Geothermal Research 201 (2011) 301–311

Fig. 6. Cone N3, erupted 1975 (August 2007). looking (VNIR3N) and backward-looking (VNIR3B) scenes of the allows morphometric measurements with accuracy similar to medi- same point on the surface. This creates the effect of automatic stereo um-scale topographic maps (1:50,000). correlation which is used for DEM generation with spatial resolution The procedure for deriving DEM from ASTER data began with a of 15 m. Generally, the root mean square error (RMSE) in ASTER DEM definition of the study area (Tolbachik volcano field) cropped from elevations ranges from ±7 m to ±15 m, depending on availability the entire ASTER imagery (60*60 km), for VNIR3N and VNIR3B bands. and quality of GCPs (known ground control points) and image quality The next step was orthorectification of these two bands by rational (Hirano et al., 2003). In vegetationless areas ASTER digital elevation polynomial coefficients (RPC) of sensor geometry, accordingly to models are expected to approach a better vertical accuracy (about image metadata and ASTER product documentation. Then DEM was 10 m) (Goncalves and Oliveira, 2004; Hirano et al., 2003) which generated by means of the ENVI 4.3 DEM extraction wizard using the

Fig. 7. Northern Fissure cinder cones, lava and tephra covered surrounding area, erupted 1975, on the background of Tolbachik volcano (August 2007). M. Inbar et al. / Journal of Volcanology and Geothermal Research 201 (2011) 301–311 307

Fig. 8. The view of the crater of Cone N1.

orthorectified VNIR3N band as a left stereo image and the orthor- between the cone's highest point and the basal isoline on the shortest ectified VNIR3B band as a right one. In this process RPC are used to horizontal distance). generate tie points and to calculate the stereo image pair relationship. The availability and accuracy of ground control points that can be 6. Results located precisely on ASTER stereo pairs determine the accuracy of the absolute orientation. Accordingly, we used field GPS measurements of Morphometric parameters of nine cinder cones on the Tolbachik 40 locations covering nine cinder cones as ground control points to tie volcano field were calculated from ASTER DEM and topographic maps the horizontal and vertical reference systems to UTM geodetic (Table 1). All these cinder cones are higher than 50 m and could be coordinates. As a result absolute horizontal and vertical orientation easily identified on the landscape by visual examination of a three- was achieved. The final DEM was generated at spatial resolution of dimensional view of DEM (Fig. 11). For all observed cinder cones the 30 m. Holes in the generated DEM were filled by automated volume values (in km3) were derived from ASTER DEM. interpolation, and then smoothed by 3×3 low-pass filter to reduce The studied cinder cones were divided into two groups, according the effect of possible artifacts caused by enhanced recognition of to age of eruption. The youthful cones formed during the last topographic features. For visualization of the obtained DEM (Fig. 10), Tolbachik eruption (1975–1976) and the older ones during last isolines were generated and then each studied cinder cone was 1500–1700 years which were dated by tephrochronology and 14C. The topographically separated from the surroundings, accordingly to basal field GPS measurements of these cones show the cross-section isoline. elevation profiles (Fig. 4). Some morphologically complex cones in Definition of the base of a cinder cone allows for the separation of ASTER DEM (with fewer GCPs) were subject to error, due to excessive each cinder cone from its topographic vicinity. In our study the base of interpolation (e.g. Cone N3) or because of blunder artifacts (e.g. the cinder cone was allocated by the lowest elevation isoline with an Krasniy cone). Fig. 11 shows a general view of the observed cinder average of slope pixels more than 5°. The basal plane elevation was cones (accordingly to ASTER DEM) which vary in size, morphology estimated as the average elevation value of all DEM pixels located complexity and the volume of pyroclastic material (Table 1). along the basal outline. Then we calculated the slope of the cinder The cones height value derived from the ASTER DEM and from the cone as the average of pixel slope values situated along the steepest topographic maps are similar, ranging from ~50 to ~240 m. The main profile with the greatest elevation difference (cone height, h) morphometric parameters of slope and height/base ratio according to between the highest cone point and the basal outline on the shortest ASTER DEM (30 m of grid cell) prove significant correlation with old and horizontal distance. The basal isoline was fitted to an ellipse shape by new cone age groups (Table 1). In all studied cones of ASTER DEM the means of an ad-hoc developed IDL program, to facilitate the width highest slope values correspond to the recent cones (Table 1). The measurements. Then the cone width (W) was calculated as the average slope for the recent cones group is 31.9° (ASTER DEM) and 31.7° average of the maximum and minimum axes of this ellipse. (MAP); the “old” cone group shows considerably smaller average slope The slope is usually defined as a plane tangent to a topographic values: 26.3° and 27.1°, respectively. Association of high values with surface of the DEM at a given pixel (Burrough, 1986). To derive the modern cones is also evident in h/W ratio (except Cone N3 and Cone slope information from the ASTER DEM, the topographic modeling Krasniy). Average values of height–base ratio are: 0.22 (ASTER) and 0.21 procedure of ENVI 4.3 was performed to produce slope values in (maps) for recent cones; 0.17 (ASTER) and 0.16 (MAP) for older cones. A degrees for every pixel in the DEM. Thus, the steepest slope of the possible reason for discrepancy of h/W ratio of Cone N3 and Cone cinder cone was calculated as an average of pixel slope values for the Krasniy might be the comparatively complex morphology of these profile with the highest run/rise ratio (largest elevation difference constructions, which have an effect on the measurements. By definition 308 M. Inbar et al. / Journal of Volcanology and Geothermal Research 201 (2011) 301–311

Fig. 9. (a) Three GFTE cones of the Northern fissure on the 1:25,000 topographic map; (b) the perspective view of three GFTE cones of the Northern fissure in ASTER DEM.

ASTER data have lesser spatial and height accuracy than the topography map; nevertheless in our case similar morphometric characteristic were obtained. This finding also verifies the reliability of DEM-based volume calculations. In the “old” cone group the cones of smaller volume have been associated with smaller h/W ratio. The h/W ratio reflects the cone's degradation rate, decreasing with age, as material is removed from the crest to the cone base. For recent GFTE cones the obtained h/W values were above 0.18 and actually reached 0.26 for Cone N2 according to ASTER DEM measurements (0.24 according to the topographic map). The average h/W ratio of cinder cones dated as several thousands of Fig. 10. The DEM (Cone N1) with over draped isolines. years was about 0.15 for DEM and maps measurements. M. Inbar et al. / Journal of Volcanology and Geothermal Research 201 (2011) 301–311 309

Table 1 Morphometric parameters of nine cinder cones of Tolbachik volcano field measured from ASTER DEM and topography maps.

Volume, Cone height (h), m Cone slope (s), Mean basal Height–base ratio km3 degrees diameter (D), m (h/D)

ASTER MAP ASTER MAP ASTER MAP ASTER MAP

Recent cones 1 Cone N1 0.135 239 228 32.4 32.0 1125 1050 0.21 0.22 2 Cone N2 0.097 205 211 32.7 31.8 800 870 0.26 0.24 3 Cone N3 0.017 97 95 31.5 32.0 533 525 0.18 0.18 4 Cone Yuzhniy 0.016 117 113 31.2 31.0 550 550 0.21 0.21 “Old” Cones 5 Cone Krasniy 0.086 202 138 29.1 29.5 810 700 0.25 0.20 6 Cone Alaid 0.118 226 230 28.8 32.0 1409 1325 0.16 0.17 7 Cone “Stariy” 0.120 96 107 26.9 27.7 901 825 0.12 0.13 8 Cone “Camp” 0.009 56 70 25.2 23.5 425 500 0.13 0.14 9 Cone Kamenistaya 0.017 107 109 21.6 23.0 603 650 0.18 0.17

Despite the appearance of small-scale artifacts (caused by stereo where vertical error often appeared about three times greater than pair matching) the concentration of field GCPs over the studied cones horizontal. led to their realistic representation in ASTER DEM (Fig. 11). The spatial accuracy of ASTER DEM has been evaluated by comparison of DEM 7. Discussion value to field ground control points (Table 2). The evaluation of horizontal and vertical root mean square error (RMSE) of each cinder The evolution of the Tolbachik volcanic landscape, 31 years after cone has demonstrated acceptable level of accuracy, typical for DEMs cessation of the volcanic activity, with a number of recent monoge- derived from ASTER data. In all cinder cones vertical errors were netic cinder cones offers the opportunity to learn about the significantly higher than horizontal, especially among old cones development of the landscape, a central issue in geomorphologic

Fig. 11. Digital elevation models of nine cinder cones studied: a — Cone N1; b — Cone N2; c — Cone N3; d — Yuzniy; e — Kamenistaya; f — Cone “Camp”;g— Cone Stariy; h — Cone Alaid and i — Cone Krasniy. Note the visual difference in the shape of new (a–d) and old cones (e–i). 310 M. Inbar et al. / Journal of Volcanology and Geothermal Research 201 (2011) 301–311

Table 2 affected by subjective decisions of the analyst. Thus, the straightfor- Distribution of horizontal and vertical RMS errors of the ASTER DEM for the nine cinder ward numerical comparison of morphometric values calculated from cones studied. DEM and topographic maps might be erroneous. In such a case, RMSE Vertical, m RMSE Horizontal, m calculated morphometric parameters might be affected by smoother Recent cones 1 Cone N1 9.9 14.6 representation of the surface in coarse-resolution DEMs. Similarly, the 2 Cone N2 9.6 14.8 comparison of DEMs of different spatial resolution when younger 3 Cone N3 8.4 13.1 cinder cones depicted on a coarse-resolution DEM might result in the 4 Cone Yuzhniy 8.3 12.2 similar morphometric values as older cones calculated from a high- “Old” Cones 5 Cone Krasniy 15.1 23.3 6 Cone Alaid 10.6 16.9 resolution DEM. For correct morphometric dating of the cinder cones 7 Cone “Stariy” 9.0 12.9 series we suggest the comparison of morphometric parameters 8 Cone “Camp” 9.3 13.7 systematically calculated from the one source of elevation data. 9 Cone Kamenistaya 4.1 6.1 The accuracy of ASTER DEM has demonstrated typical spatial error for DEMs derived from ASTER data (Table 2) which we found suitable for morphometric measurements on small-scale pyroclastic construc- studies. The h/D ratio values for the recent cones as high as 0.21 tions like cinder cones. (ASTER DEM) or 0.27 (MAP), are consistent with the published data The influence of climate is reflected in the highest rates of erosion for several areas in the world in different climatic conditions (Inbar of the area, due to the harsh weather conditions. The slope angle and Risso, 2001). There is a clear relationship between slope and h/W decreases from an initial value of 31°–34° for the 1975 cones to ratio, according to Woods (1980b) predicted curve (Fig. 12) The 25°–29° for the cones 1500–2000 years BP. The Paricutín cone in Tolbachik cones have high slope values like the Paricutin, Navidad and Mexico erupted in 1943 had an initial slope of 31° to 33° (Segerstrom, other new cones (Moreno and Gardeweg, 1989; Inbar et al., 1994). 1950) declining slightly to 32° and 31°, 45 years later (Inbar et al., This paper assessed the capability of remotely sensed ASTER DEM 1994). There is a rapid decline of the slope values in the first stage of for morphometric measurements of cinder cones on the example of erosion. Recent cones on Mt Etna show a similar slope decrease of 10 the Tolbachik volcanic field in Kamchatka. The morphometric degrees in only 450 years (Wood, 1980b). Cone evolution shows parameters of cone slope, height and width were calculated from erosional rounding of rim and deposition at base, in accordance with the remotely sensed DEMs and compared with the morphometric linearly slope-dependent transport models (Pelletier and Cline, 2007). parameters measured from the topographic map (Table 1). The measured morphometric parameters of the recent Tolbachik Morphometric parameters were calculated for the nine cinder cinder cones are important for determining erosion rates, as the cones studied, divided into two age groups of recent and older cones. measured erosion values for the Paricutin cone after the 1943 ASTER DEM and topographic map were compared to assess different eruption (Segerstrom, 1950) and the Navidad cone erupted in methods on morphometric measurements of height, width, h/W ratio December 1988 in (Moreno and Gardeweg, 1989). A comparison and slope. The morphometric parameters calculated from ASTER DEM with the cinder cones older by several millennia in the area shows and topographic map are numerically different from each other but that the climatic conditions and the slow development of vegetation well correlated with age groups of cinder cones. For example, cover determine high rates of erosion and rapid change in morpho- calculated from both ASTER DEM and topographic map, all the four metric values. Variability in morphometric values of the recent cinder recent cones have the highest slope values; the two cones with the cones is due to their lithological coarse composition. highest slopes (Cones N1 and N2) correlate with the highest values of Peaks of erosion occurred probably in the first stage of 1 or 2 years h/W ratio (Table 1). after the eruption, with the stripping of the fine ash material, as The comparison of morphometric parameters calculated from recorded in several volcanic areas of the world (Swanson et al., 1983). ASTER DEM and topographic map clearly supports the concept of Accelerated peaks of erosion were also measured after forest fires in relative morphometric dating as the most recent cinder cones are non vegetated areas (Inbar et al., 1998). always associated with the highest slopes and h/W ratio (Table 1). The morphometric parameters of the recent Tolbachik cinder However, in DEM spatial resolution controls the ability to represent cones serve as a valuable benchmark data for determining erosion local topography of cinder cones by interpolation of elevation data rates and future research on the geomorphological evolution of points in the vicinity of breaks in slope. From other hand, the volcanic landscape. morphometric measurements from topographic maps might be The erosional processes in the climatic conditions of the Kamchatka peninsula are more intensive than the published values for other cinder cones in volcanic fields (e.g. the subtropical climate of the Mexican Volcanic field (Hooper, 1995), the semiarid climate of Payun Matru (Inbar and Risso, 2001), or the of the Golan plateau (Inbar et al., 2008) showing measurable morpholog- ical changes in the range of thousands of years.

Acknowledgments

The research was supported by the Ministry of Science of Israel grant 3-3572 and the Russian Foundation for Basic Research. The authors are grateful to the Institute of Volcanology and Seismology in Petropavlosk- Kamchatsky, Russia for logistic support of the study. We thank the two reviewers for their constructive comments on the manuscript.

Fig. 12. Cone slope values and Hco/Wco ratio values for cinder cones in the Payun Matru References Volcanic Field, Navidad cone and Tolbachik cones. Predicted curve is from weathering and mast wasting model by Wood (1980b). Note that numeration of Kamchatka cones Bloomfield, K., 1975. A late-Quaternary monogenetic volcanic field in central Mexico. conforms their order in Tables 1 and 2. Geologischen Rundschau 64, 476–497. M. Inbar et al. / Journal of Volcanology and Geothermal Research 201 (2011) 301–311 311

Braitseva, O.A., Melekestsev, I.V., Litasova, S.N., Sulerzhitsky, L.D., Ponomareva, V.V., Inbar, M., Tamir, M., Wittenberg, L., 1998. Runoff and erosion processes after a forest fire 1993. Radiocarbon dating and tephrochronology in Kamchatka. Radiocarbon 35 in Mt. Carmel, a Mediterranean area. Geomorphology 24, 17–34. (3), 463–477. Inbar, M., Gilichinsky, M., Melekestsev, I., Melnikov, D., 2008. A Morphological and Braitseva, O.A., Ponomareva, V.V., Sulerzhitsky, L.D., Melekestsev, I.V., Bailey, J., 1997. Morphometric Study of Cinder Cones in Kamchatka and . Proceedings Holocene key-marker tephra layers in Kamchatka, Russia. Quaternary Research 47, Israel Geological Annual Meeting, Nazareth, Israel, p. 44. 125–139. Kervyn, M., Ernst, G.G.J., Goossens, R., Jacobs, P., 2008. Mapping volcano topography Burrough, P.A., 1986. Principles of Geographical Information Systems for Land with remote sensing: ASTER vs. SRTM. International Journal of Remote Sensing 29 Resources Assessment. Clarendon Press, Oxford. pp. 147–166. (22), 6515–6538. Dirksen, O.V., Melekestsev, I.V., 2000. Chronology, evolution and morphology of plateau Martin del Pozzo, A.L., 1982. Monogenetic vulcanism in Sierra Chichinautzin, Mexico. eruptive centers in Avacha River Area, Kamchatka. Vulkanologia i Seismologia 21, Bulletin of Volcanology 45, 9–24. 617. Moreno, H., Gardeweg, M.C., 1989. La erupcion reciente en el complejo volcanico Dohrenwend, J.C., Wells, S.G., Turrin, B.D., 1986. Degradation of Quaternary cinder Lonquimay (Diciembre 1988), del Sur. Revista Geologica de Chile 16,1, 93–117. cones in the Cima volcanic field, , . Geological Society of Parrot, J.F., 2007. Tri-dimensional parameterization: an automated treatment to study America Bulletin 97, 421–427. the evolution of volcanic cones. Geomorphologie 3, 247–258. Dóniz, J., Romero, C., Coello, E., Guillén, C., Sánchez, N., García-Cacho, L., García, A., 2008. Pelletier, J.D., Cline, M.L., 2007. Nonlinear slope-dependent sediment transport in Morphological and statistical characterization of recent mafic volcanism on cinder cone evolution. 35 (12), 1067–1070. Tenerife (Canary Islands, Spain). Journal of Volcanology and Geothermal Research Ponomareva, V.V., Sulerzhitsky, L.D., Dirksen, O.V., Zaretskaya, N.E., 2001. Holocene 173, 185–195. paleosols as records of intervals of volcanic quiescence in the Kuril Lake , South Dvigalo, V.N., Selezniov, B.V., Magus'kin, M.A., 1980. New results of aerial photo analysis Kamchatka. , Chronology and Archeology. CDERAD Press, pp. 91–100. of the Great Fissure Tolbachik Eruption. Volcanology and Seismology 3, 90–93. Porter, S.C., 1972. Distribution, morphology and size frequency of cinder cones on Fedotov, S.A., Masurenkov, Y.P., 1991. Active Volcanoes of Kamchatka, vol. 1. Nauka volcano, Hawaii. Geological Society of America Bulletin 83, 3607–3612. Publishers, Moscow, p. 21. Scott, D.H., Trask, N.J., 1971. Geology of the Luna Crater volcanic field, Nye County, Flerov, G.B., Andreev, V.N., Budnikov, V.A., Tsurupa, A.I., 1984. Petrology of eruptive . USGS Prof. Paper, 599-I. 22 pp. products. Great Tolbachik fissure eruption. Kamchatka, 1975–1976. Nauka, Segerstrom, K., 1950. Erosion studies at Paricutin, state of Michoacan, Mexico. USGS Moscow, pp. 223–284. Bull. 965-A 164 pp. Goncalves, J.A., Oliveira, A.M., 2004. Accuracy Analysis of DEMS derived from ASTER Settle, M., 1979. The structure and emplacement of cinder cone fields. American Journal Imagery. International Archives of Photogrammetry and Remote Sensing 35, of Science 279, 1089–1107. 168–172. Solomina, O., Wiles, G., Shiraiwa, T., D'Arrigo, R., 2007. Multiproxy records of climate Hasenaka, T., Carmichael, I., 1985. The cinder cones of Michoacán-Guanajuato, central variability for Kamchatka for the past 400 years. Climate of the Past 3, 119–128. Mexico: their age, volume, distribution and discharge rate. Journal of Swanson, F.J., Collins, B., Dunne, T., Wicherski, B.P., 1983. Erosion of tephra from hillslopes Volcanology and Geothermal Research 25, 104–124. near Mount St. Helens and other volcanoes. Symp. on Erosion Control in Volcanic Hirano, A., Welch, R., Lang, H., 2003. Mapping from ASTER stereo image data: DEM Areas, Seattle, July 1982. Public Works Research Institute, Ibaraki, Japan, pp. 183–221. validation and accuracy assessment. ISPRS Journal of Photogrammetry and Remote Syrin, A.N., 1968. Regional and Local Volcanism Relationship. Nauka, Moscow. 149 pp. Sensing 57, 356–370. (in Russian). Hooper, D.M., 1995. -simulation models of scoria cone degradation in the Thouret, J.C., 1999. Volcanic geomorphology — an overview. Earth Sciences Reviews 47, Colima and Michoacan-Guanajuato volcanic fields, Mexico. Geofísica Internacional 95–131. 34, 321–340. Wolfe, E.W., Ulrich, G.E., Newhall, C.G., 1987. Geologic Map of the Northwest Part of the Hooper, D.M., Sheridan, M.F., 1998. Computer-simulation models of scoria cone San Francisco Volcanic Field, North-central Arizona. USGS Misc. Field Stud. Map, MF degradation. Journal of Volcanology and Geothermal Research 83, 241–287. 1957. Inbar, M., Risso, C., 2001. A morphological and morphometric analysis of a high density Wood, C.A., 1980a. Morphometric evolution of cinder cones. Journal of Volcanology and cinder cone volcanic field — Payun Matru, south-central Andes, Argentina. Geothermal Research 7, 387–413. Zeitschrift für Geomorphologie 45, 321–343. Wood, C.A., 1980b. Morphometric analysis of cinder cone degradation. Journal of Inbar, M., Lugo Hubp, J., Villers Ruiz, L., 1994. The geomorphological evolution of the Volcanology and Geothermal Research 8, 137–160. Paricutin cone and lava flows, Mexico, 1943–1990. Geomorphology 57–76.